import numpy as np
import pandas as pd
import plotly.graph_objects as go

Załadowanie zbioru danych:

df = pd.read_csv('complete.csv')
df.head()

	awardYear	category	categoryFullName	sortOrder	portion	prizeAmount	prizeAmountAdjusted	dateAwarded	prizeStatus	motivation	...	org_founded_country	org_founded_countryNow	org_founded_locationString	ind_or_org	residence_1	residence_2	affiliation_1	affiliation_2	affiliation_3	affiliation_4
0	2001	Economic Sciences	The Sveriges Riksbank Prize in Economic Scienc...	2	1/3	10000000	12295082	2001-10-10	received	for their analyses of markets with asymmetric ...	...	NaN	NaN	NaN	Individual	NaN	NaN	Stanford University, Stanford, CA, USA	NaN	NaN	NaN
1	1975	Physics	The Nobel Prize in Physics	1	1/3	630000	3404179	1975-10-17	received	for the discovery of the connection between co...	...	NaN	NaN	NaN	Individual	NaN	NaN	Niels Bohr Institute, Copenhagen, Denmark	NaN	NaN	NaN
2	2004	Chemistry	The Nobel Prize in Chemistry	1	1/3	10000000	11762861	2004-10-06	received	for the discovery of ubiquitin-mediated protei...	...	NaN	NaN	NaN	Individual	NaN	NaN	Technion - Israel Institute of Technology, Hai...	NaN	NaN	NaN
3	1982	Chemistry	The Nobel Prize in Chemistry	1	1	1150000	3102518	1982-10-18	received	for his development of crystallographic electr...	...	NaN	NaN	NaN	Individual	NaN	NaN	MRC Laboratory of Molecular Biology, Cambridge...	NaN	NaN	NaN
4	1979	Physics	The Nobel Prize in Physics	2	1/3	800000	2988048	1979-10-15	received	for their contributions to the theory of the u...	...	NaN	NaN	NaN	Individual	NaN	NaN	International Centre for Theoretical Physics, ...	Imperial College, London, United Kingdom	NaN	NaN

5 rows × 52 columns

Stworzenie kolumny zawierającej wiek laureatów w momencie odebrania nagrody (dla uproszczenia liczony rocznikowo):

df = df[ ~(df['birth_date'].isna()) ]
df['birth_year'] = df['birth_date'].str.split('-').apply(lambda x: int(x[0]))
df['age'] = df['awardYear'] - df['birth_year']

Funkcja dzieląca zbiór danych na pozbiory pogrupowane po kategoriach (plus dane dla wszystkich ogółem):

def split_data_by_category(data, categories):
    return [data['age']] + [data[data['category'] == category]['age'] for category in categories]

Generowanie wykresu:

df_splitted = split_data_by_category(df, df['category'].unique())

# podstawowy wykres
fig = go.Figure()
fig.update_layout(title = "Age of Nobel Laureates")
fig.add_trace(go.Box(y = df['age'], name = 'All'))
for y, name in zip(df_splitted, df['category'].unique()):
    fig.add_trace(go.Box(y = y, name = name))

# dodanie możliwości zmieniania na przed i po 1969
fig.update_layout(
    updatemenus=[
        dict(
            active=0,
            buttons=list([
                dict(label="All years",
                     method="update",
                     args=[{"y": df_splitted},
                           {"title": "Age of Nobel Laureates"}]),
                dict(label="Since 1969",
                     method="update",
                     args=[{"y": split_data_by_category(df[df['awardYear'] >= 1969], categories = df['category'].unique())},
                           {"title": "Age of Nobel Laureates since 1969"}]),
                dict(label="Before 1969",
                     method="update",
                     args=[{"y": split_data_by_category(df[df['awardYear'] < 1969], categories = df['category'].unique())},
                           {"title": "Age of Nobel Laureates before 1969"}])
            ])
        )
    ])